Page MenuHomePhabricator
Files F31839892

KkConverter.php
MuratKaribay (MuratKaribay)
Actions

Authored By
MuratKaribay
May 25 2020, 9:39 AM
Size
10 KB
Referenced Files
None
Subscribers
None

KkConverter.php
View Options

<?php
require_once( dirname(__FILE__).'/../LanguageConverter.php' );
require_once( dirname(__FILE__).'/LanguageKk_cyrl.php' );
define( 'KK_C_UC', 'АӘБВГҒДЕЁЖЗИЙКҚЛМНҢОӨПРСТУҰҮФХҺЦЧШЩЪЫІЬЭЮЯ' ); # Kazakh Cyrillic uppercase
define( 'KK_C_LC', 'аәбвгғдеёжзийкқлмнңоөпрстуұүфхһцчшщъыіьэюя' ); # Kazakh Cyrillic lowercase
define( 'KK_L_UC', 'AÄBCÇDEFGĞHİIJKLMNŊOÖPQRSŞTUÜVWXYZ' ); # Kazakh Latin uppercase
define( 'KK_L_LC', 'aäbcçdefgğhiıjklmnŋoöpqrsştuüvwxyz' ); # Kazakh Latin lowercase
define( 'KK_A', 'ءابپتجحدرزسشعفقكلمنڭەوۇۋۆىيچھ' ); # Kazakh Arabic
define( 'HAMZA', 'ء' ); # U+0621 ARABIC LETTER HAMZA
/** Kazakh (Қазақша)
* converter routines
*
* @ingroup Language
*/
class KkConverter extends LanguageConverter{
function __construct($langobj,$maincode,
$variants=array(),
$variantfallbacks=array(),
$flags = array()){
parent::__construct($langobj,$maincode,
$variants,$variantfallbacks,$flags);
$this->loadRegs();
}
function loadDefaultTables(){
$kk2Cyrl = array();
$kk2Latn = array();
$kk2Arab = array();
$this->mTables = array(
'kk-cyrl' => new ReplacementArray($kk2Cyrl),
'kk-latn' => new ReplacementArray($kk2Latn),
'kk-arab' => new ReplacementArray($kk2Arab),
'kk' => new ReplacementArray()
);
}
function loadRegs(){
$this->mCyrl2Latn = array(
'/^И(['.KK_C_UC.']|$)/u' => 'İY$1', '/^И(['.KK_C_LC.']|$)/u' => 'İy$1',
'/^Ю(['.KK_C_UC.']|$)/u' => 'İYW$1', '/^Ю(['.KK_C_LC.']|$)/u' => 'İyw$1',
'/^Я(['.KK_C_UC.']|$)/u' => 'İYA$1', '/^Я(['.KK_C_LC.']|$)/u' => 'İya$1',
'/Ё/u' => 'Ö', '/ё/u' => 'ö',
'/И/u' => 'İY', '/и/u' => 'iy',
'/Ю/u' => 'İYW', '/ю/u' => 'iyw',
'/Я/u' => 'İYA', '/я/u' => 'iya',
'/У/u' => 'Ú', '/у/u' => 'ú',
'/Щ/u' => 'Ş', '/щ/u' => 'ş',
'/[ъЪ]/u' => '', '/[ьЬ]/u' => '',
'/А/u' => 'A', '/а/u' => 'a', '/Ә/u' => 'Ä', '/ә/u' => 'ä',
'/Б/u' => 'B', '/б/u' => 'b', '/В/u' => 'V', '/в/u' => 'v',
'/Г/u' => 'G', '/г/u' => 'g', '/Ғ/u' => 'Ğ', '/ғ/u' => 'ğ',
'/Д/u' => 'D', '/д/u' => 'd', '/Е/u' => 'E', '/е/u' => 'e',
'/Ё/u' => 'Ö', '/ё/u' => 'ö', '/Ж/u' => 'J', '/ж/u' => 'j',
'/З/u' => 'Z', '/з/u' => 'z', '/Й/u' => 'Y', '/й/u' => 'y',
'/К/u' => 'K', '/к/u' => 'k', '/Қ/u' => 'Q', '/қ/u' => 'q',
'/Л/u' => 'L', '/л/u' => 'l', '/М/u' => 'M', '/м/u' => 'm',
'/Н/u' => 'N', '/н/u' => 'n', '/Ң/u' => 'Ŋ', '/ң/u' => 'ŋ',
'/О/u' => 'O', '/о/u' => 'o', '/Ө/u' => 'Ö', '/ө/u' => 'ö',
'/П/u' => 'P', '/п/u' => 'p', '/Р/u' => 'R', '/р/u' => 'r',
'/С/u' => 'S', '/с/u' => 's', '/Т/u' => 'T', '/т/u' => 't',
'/Ұ/u' => 'U', '/ұ/u' => 'u', '/Ү/u' => 'Ü', '/ү/u' => 'ü',
'/Ф/u' => 'F', '/ф/u' => 'f', '/Х/u' => 'H', '/х/u' => 'h',
'/Һ/u' => 'H', '/һ/u' => 'h', '/Ц/u' => 's', '/ц/u' => 's',
'/Ч/u' => 'Ç', '/ч/u' => 'ç', '/Ш/u' => 'С', '/ш/u' => 'с',
'/Щ/u' => 'C', '/щ/u' => 'C', '/Ы/u' => 'I', '/ы/u' => 'ı',
'/І/u' => 'İ', '/і/u' => 'i', '/Э/u' => 'E', '/э/u' => 'e',
//Cyrillic -> Latin for letters И and У
'/ИІ/ui' => 'İYİ', '/иі/ui' => 'iyi',
'/ИЫ/ui' => 'IYI', '/иы/ui' => 'ıyı',
'/ИЯ/ui' => 'İYA', '/ия/ui' => 'iya',
);
$this->mCyrl2Arab = array(
## Punctuation -> Arabic
'/#|№|No\./u' => '؀', # &#x0600;
'/\,/' => '،', # &#x060C;
'/;/' => '؛', # &#x061B;
'/\?/' => '؟', # &#x061F;
'/%/' => '٪', # &#x066A;
'/\*/' => '٭', # &#x066D;
## Cyrillic -> Arabic
'/[еэ]/ui' => 'ە', '/[ъь]/ui' => '',
'/[аә]/ui' => 'ا', '/[оө]/ui' => 'و', '/[ұү]/ui' => 'ۇ', '/[ыі]/ui' => 'ى',
'/[и]/ui' => 'ي', '/ё/ui' => 'يو', '/ю/ui' => 'يۋ', '/я/ui' => 'يا', '/[й]/ui' => 'ي',
'/ц/ui' => 'س', '/щ/ui' => 'ش',
'/һ/ui' => 'ح', '/ч/ui' => 'چ',
'/б/ui' => 'ب', '/в/ui' => 'ۆ', '/г/ui' => 'گ', '/ғ/ui' => 'ع',
'/д/ui' => 'د', '/ж/ui' => 'ج', '/з/ui' => 'ز', '/к/ui' => 'ك',
'/қ/ui' => 'ق', '/л/ui' => 'ل', '/м/ui' => 'م', '/н/ui' => 'ن',
'/ң/ui' => 'ڭ', '/п/ui' => 'پ', '/р/ui' => 'ر', '/с/ui' => 'س',
'/т/ui' => 'ت', '/у/ui' => 'ۋ', '/ф/ui' => 'ف', '/х/ui' => 'ح',
'/ш/ui' => 'ش',
// Cyrillic -> Arabic MOJE UPRAVY - upravil som yya = ya
'/ия/ui' => 'يا',
);
}
function parseManualRule($rule,$flags=array()){
if(in_array('T',$flags)){
return parent::parseManualRule($rule,$flags);
}
// otherwise ignore all formatting
foreach($this->mVariants as $v){
$carray[$v] = $rule;
}
return $carray;
}
/*
* A function wrapper:
* - if there is no selected variant, leave the link
* names as they were
* - do not try to find variants for usernames
*/
function findVariantLink(&$link, &$nt, $ignoreOtherCond = false){
// check for user namespace
if(is_object($nt)){
$ns = $nt->getNamespace();
if($ns==NS_USER || $ns==NS_USER_TALK)
return;
}
$oldlink=$link;
parent::findVariantLink($link, $nt, $ignoreOtherCond);
if($this->getPreferredVariant()==$this->mMainLanguageCode)
$link=$oldlink;
}
/*
* An ugly function wrapper for parsing Image titles
* (to prevent image name conversion)
*/
function autoConvert($text, $toVariant=false){
global $wgTitle;
if(is_object($wgTitle) && $wgTitle->getNameSpace()==NS_FILE){
$imagename = $wgTitle->getNsText();
if(preg_match("/^$imagename:/",$text)) return $text;
}
return parent::autoConvert($text,$toVariant);
}
/**
* It translates text into variant
*/
function translate($text, $toVariant){
global $wgContLanguageCode;
$text = parent::translate($text, $toVariant);
$letters = '';
switch($toVariant){
case 'kk-cyrl':
$letters = KK_L_UC . KK_L_LC . 'ʺʹ#0123456789';
$wgContLanguageCode = 'kk';
break;
case 'kk-latn':
$letters = KK_C_UC . KK_C_LC . '№0123456789';
$wgContLanguageCode = 'kk-Latn';
break;
case 'kk-arab':
$letters = KK_C_UC . KK_C_LC . 'ʺʹ'.*/',;\?%\*№0123456789';
$wgContLanguageCode = 'kk-Arab';
break;
default:
$wgContLanguageCode = 'kk';
return $text;
}
// disable conversion variables like $1, $2...
$varsfix = '\$[0-9]';
$matches = preg_split('/' . $varsfix . '[^' . $letters . ']+/u', $text, -1, PREG_SPLIT_OFFSET_CAPTURE);
$mstart = 0;
$ret = '';
foreach( $matches as $m ){
$ret.= substr($text, $mstart, $m[1]-$mstart);
$ret.= $this->regsConverter($m[0], $toVariant);
$mstart = $m[1] + strlen($m[0]);
}
return $ret;
}
function regsConverter($text, $toVariant){
if ($text == '') return $text;
$pat = array();
$rep = array();
switch($toVariant){
case 'kk-arab':
$letters = KK_C_LC.KK_C_UC/*.KK_L_LC.KK_L_UC*/;
$front = 'әөүіӘӨҮІ'/*.'äöüiÄÖÜİ'*/;
$excludes = 'еэгғкқЕЭГҒКҚ'/*.'egğkqEGĞKQ'*/;
// split text to words
$matches = preg_split( '/[\b\s\-\.:]+/', $text, -1, PREG_SPLIT_OFFSET_CAPTURE);
$mstart = 0;
$ret = '';
foreach($matches as $m){
$ret.= substr($text, $mstart, $m[1] - $mstart);
// is matched the word to front vowels?
// exclude a words matched to е, э, г, к, к, қ,
// them should be without hamza
if(preg_match('/['.$front.']/u', $m[0]) && !preg_match('/['.$excludes.']/u', $m[0])){
$ret .= preg_replace('/['.$letters.']+/u', HAMZA.'$0', $m[0]);
} else{$ret .= $m[0];}
$mstart = $m[1] + strlen($m[0]);
}
$text =& $ret;
foreach($this->mCyrl2Arab as $pat => $rep){
$text = preg_replace($pat, $rep, $text);
}
return $text;
break;
case 'kk-latn':
foreach($this->mCyrl2Latn as $pat => $rep){
$text = preg_replace($pat, $rep, $text);
}
return $text;
break;
default:
return $text;
}
}
/*
* We want our external link captions to be converted in variants,
* so we return the original text instead -{$text}-, except for URLs
*/
function markNoConversion($text, $noParse=false){
if($noParse || preg_match("/^https?:\/\/|ftp:\/\/|irc:\/\//", $text))
return parent::markNoConversion($text);
return $text;
}
function convertCategoryKey($key){
return $this->autoConvert($key, 'kk');
}
}
/**
* class that handles Cyrillic, Latin and Arabic scripts for Kazakh
* right now it only distinguish kk_cyrl, kk_latn, kk_arab.
*
* @ingroup Language
*/
class LanguageKk extends LanguageKk_cyrl{
function __construct(){
global $wgHooks;
parent::__construct();
$variants = array('kk', 'kk-cyrl', 'kk-latn', 'kk-arab');
$variantfallbacks = array(
'kk' => 'kk-cyrl',
'kk-cyrl' => 'kk',
'kk-latn' => 'kk',
'kk-arab' => 'kk',
);
$this->mConverter = new KkConverter($this, 'kk', $variants, $variantfallbacks);
$wgHooks['ArticleSaveComplete'][] = $this->mConverter;
}
/**
* Work around for right-to-left direction support in kk-arab and kk-cn
*
* @return bool
*/
function isRTL(){
$variant = $this->getPreferredVariant();
if($variant == 'kk-arab'){
return true;
} else{return parent::isRTL();}
}
/*
* It fixes issue with ucfirst for transforming 'i' to 'İ'
*
*/
function ucfirst ($string){
$variant = $this->getPreferredVariant();
if (($variant == 'kk-latn') && $string[0] == 'i'){
$string = 'İ' . substr($string, 1);
} else {$string = parent::ucfirst($string);}
return $string;
}
/*
* It fixes issue with lcfirst for transforming 'I' to 'ı'
*
*/
function lcfirst ($string){
$variant = $this->getPreferredVariant();
if (($variant == 'kk-latn') && $string[0] == 'I'){
$string = 'ı' . substr($string, 1);
} else {$string = parent::lcfirst($string);}
return $string;
}
function convertGrammar($word, $case){
wfProfileIn( __METHOD__ );
$variant = $this->getPreferredVariant();
switch ($variant){
case 'kk-arab':
$word = parent::convertGrammarKk_arab($word, $case);
break;
case 'kk-latn':
$word = parent::convertGrammarKk_latn($word, $case);
break;
case 'kk-cyrl':
case 'kk':
default:
$word = parent::convertGrammarKk_cyrl($word, $case);
}
wfProfileOut(__METHOD__);
return $word;
}
}

File Metadata

Mime Type
text/x-php
Storage Engine
blob
Storage Format
Raw Data
Storage Handle
8417657
Default Alt Text
KkConverter.php (10 KB)

Event Timeline

Content licensed under Creative Commons Attribution-ShareAlike (CC BY-SA) 4.0 unless otherwise noted; code licensed under GNU General Public License (GPL) 2.0 or later and other open source licenses. By using this site, you agree to the Terms of Use, Privacy Policy, and Code of Conduct. · Wikimedia Foundation · Privacy Policy · Code of Conduct · Terms of Use · Disclaimer · CC-BY-SA · GPL